Results
Trial Identification
The search yielded 4,130 publications. The flowchart of the articles is depicted in Figure 1. One report was translated from Mandarin and one from German into English to access eligibility. Of 68 potentially eligible studies, three were excluded because they were not RCTs, 55 studies did not match the ALI or ARDS criteria, 4 trials were excluded due to fluids comparison, and 3 studies did not report the outcome investigated by this review. Detailed information on the excluded articles is listed in the Additional file 2. Finally, two trials and one subgroup from a large RCT were included in this review, and their data were analyzed.
(Enlarge Image)
Figure 1.
Data extraction flow chart. RCT, randomized controlled trial; ALI, acute lung injury; ARDS, acute respiratory distress syndrome.
Trial Characteristics
Characteristics of the three trials (the two trials and one subgroup from a large RCT that were included in this review) are shown in Table 1. Two trials were published by the same group using 25% albumin as colloid therapy and basic diuretic therapy with furosemide compared to saline in patients with ALI. The study of saline versus albumin fluid evaluation (SAFE trial) used 4% albumin compared to saline.
Risk of Bias
The Cochrane risk of bias tool is shown in Table 2, whereby risk of bias was assessed to be high, unclear or low. The overall risk of bias was unclear-to-high in the analyzed trials.
Mortality
Two trials reported 30-day mortality and the 28-day mortality was reported for the subgroup of ARDS patients in the SAFE study (Figure 2). Albumin therapy did not significantly influence either 30-day mortality alone (albumin, 10 patients out of 39 (25.6%) versus control 12 patients out of 38 (31.6%), RR = 0.81, 95% CI 0.41, 1.60, P = 0.548), or combined pooled mortality including the SAFE trial (albumin, 34 patients out of 100 (34.0%) versus 40 out of 104 (38.5%), RR = 0.89, 95% CI 0.63, 1.28, P = 0.539).
(Enlarge Image)
Figure 2.
Forest plot of pooled relative risk of death. RR, relative risk; SAFE, saline versus albumin fluid evaluation trial.
Oxygenation
Two trials reported change in oxygenation for ARDS patients (Figure 3). The WMD in change in PaO2/FiO2 significantly increased after albumin therapy in the first 24 h (WMD = 56 mmHg, 95% CI 47, 66, P <0.001, I = 0%) and 48 h (WMD = 62 mmHg, 95% CI 47, 77, P <0.001, I = 0%) as well as after 7 days (WMD = 20 mmHg, 95% CI 4, 36, P = 0.017, I = 0%). However, after 72 h, oxygenation did not differ between patients receiving albumin compared to crystalloids (WMD = 10 mmHg, 95% CI −3 - 23, p = 0.131, I = 86%).
(Enlarge Image)
Figure 3.
Change in PaO2/FiO2. PaO2/FiO2, ratio of arterial partial pressure of oxygen/fraction of inspired oxygen; WMD, weighted mean difference.